Male Partner Participation in Preventing Mother-to-Child Transmission of Human-Immunodeficiency Virus (PMCHT) and its Predictive Factors in Bishoftu, Central Ethiopia

Introduction

The growing prevalence of human immunodeficiency virus (HIV) has exposed the life of human being to a great risk. At the end of 2014, approximately 37 million people were infected with HIV among which 7% were children. A great proportion of this rate live in countries with a low socioeconomic status. In the same year, two million new infections were reported, nearly 220,000 cases of which had affected children. The majority of these children were residing in Sub-Saharan Africa and had acquired the infection during the pregnancy, childbearing, or breastfeeding by HIV-seropositive mothers (1).

Ethiopia is among the countries extensively affected by HIV. In addition to the overriding hetero-sexual transmissions, vertial transmission from mother to child accounts for greater than 90% of pediatric HIV cases. Mother-to-child HIV transmission results in big social difficulties by imposing a burden of orphans to society following the death of one or both of the parents due to HIV (2, 3).

The inhibition of mother-to-child AIDS transmission is one of the approaches to prevent pediatric HIV infection. Taking this fact in to account, the prevention of  mother-to-child HIV transmission (PMCHT) program has been implemented in Ethiopia since 2001. However, this target remains a challenge for the country due to the low coverage of the service, inadequate quality of the available service, and low male involvement in these programs (4).

Universally, the enhancement of male partner participation in PMCHT has been given a special importance. However, testing male spouses for HIV with the aim of preventing MCHT has challenged the majority of the countries with a low and middle socioeconomic status (5). The lack of male partner involvement in PMCHT deprives women of their partners’ care and support in coping with HIV infection, receiving antiretroviral therapy, and making appropriate infant feeding choices.

The inadequate or lack of male partner participation in PMCHT programs is one of the key factors inhibiting the wide scope coverage of PMCHT. Partner engagement is supposed to be precisely limited in various health institutions in Ethiopia and is one of the gaps adversely affecting PMCHT service outcomes in Ethiopia (6, 7).

Furthermore, partner engagement is part of the foremost challenges in the implementation of PMCHT programs in the area under investigation. For instance, although all women were counseled and tested for HIV during their antenatal care (ANC) visit, only 17% of the pregnant women were accompanied by their partners, which is certainly lower than the rate (50%) anticipated by the World Health Organization (8).

The reasons for low PMTCT uptake should be closely linked with the efforts to expand the coverage of maternal and child health services. The suboptimal participation of male partners in such programs is reflected as a bottleneck for PMCHT realization. Evidence suggests that the participation of male partners in the PMCHT interventions would positively affect the outcomes of such programs, such as the initiation of antiretroviral therapy, follow-up of the exposed infants, and infant nourishing choices by women. However, the factors accounting for reduced male involvement in Ethiopia have not been well described yet. Consequently, it is necessary to identify these factors to develop relevant strategies targeted toward the elimination of this shortcoming (9).

Despite the importance of this issue, little research has been focused on the identification of the factors affecting PMCHT programs. Moreover, these studies are limited in scope by focusing only on pregnant mothers and failing to examine the factors affecting male partner involvement in PMCHT interventions.

With this background in mind, the present study was conducted to determine the factors accounting for low male partner participation in PMCHT programs. The findings of our study
can be helpful in providing initiative recommendations that encourage male partner involvement in the program. Furthermore, our findings would facilitate designing effective strategies targeted toward the reduction of MCT of HIV to have HIV-free children.

Moreover, our study can generate data for other researchers to explore extra evidence regarding the factors associated with poor male partner engagement in PMCHT programs. The findings of this investigation could be also used as baseline information for the study context and other similar settings.

Materials and Methods

Study setting, design, and population

This community-based cross-sectional study was conducted within June 10-20, 2016 in Bishoftu town, which is located approximately 47 km from Addis Ababa, the capital city of Ethiopia. According to 2007 Ethiopian fiscal year population projection, the estimated total population of the town is 176,743 cases, 51% of which are female population among which 43.39% and 15.68% of the cases are in reproductive age group and pregnant, respectively.  

The town has nine kebeles with one general hospital, three public health centers, twenty
six private clinics, two non-government organization clinics, seven pharmacies, five drug stores, and one drug vender. Out of these health facilities, the general hospital, the three health centers, and one of the private clinics provide antenatal care (ANC)/PMCHT services (8). The study population corresponded to a group of male partners of reproductive age women (15-49 years) who gave birth during the previous year in Bishoftu town. The women who were in the first or second trimester of pregnancy and those with dead or seriously ill partners were excluded from the study.

Operational Definitions

In this study, the male partner involvement in ANC/PMTCHT intervention was defined as high and low, if the subject responded to 4-6 and 0-3 questions correctly (out of six questions), respectively. Furthermore, the respondent was considered to have high and low levels of knowledge about PMCHT programs if correctly answering ≥ 60% and < 60% of the knowledge items, respectively (10). Furthermore, male partners were regarded to have a positive attitude towards PMCHT programs if the attitude measurement score exceeded or equaled to the mean score of a 10-item 5-point Likert scale ranging from ‘strongly agree’ to ‘strongly disagree’; otherwise, they were categorized as having negative attitudes toward this practice.

Sample size and selection procedure

The sample size was estimated as 421 cases using the formula for single population proportion considering 95% confidence interval, 53% proportion of male involvement in PMCHT based on a similar study (11), 5% margin of error, and 10% non-response rate. Prior to data collection, a survey was conducted in all nine kebeles in the town on the male spouses whose wives had visited the facilities delivering the ANC/PMCHT services in the preceding year to prepare a sampling frame. Then, the subjects were selected from nine kebeles in proportion to the size of the study population using the simple random sampling method by means of computer-generated random numbers.

Study variables

The major outcome variable was male partner involvement in the PMCHT programs. The demographic data recorded for each respondent included age, gender, occupation, and monthly household income. In addition, data were recorded about the male partners’ knowledge about PMCHT programs, awareness regarding the provision of PMCHT services in the public health facilities, and attitude towards these services.

Data collection tool, procedure, and quality control

The questionnaire consisted of five categories, namely socio-demographic characteristics, knowledge about PMCHT programs, attitude towards ANC/PMCT services, inter-spousal communication (e.g., discussion on HIV counseling and testing), fear of stigma and discrimination, PMCHT information access, socio-cultural factors (e.g., male partner’s opinion/perceptions on attending ANC/PMCT programs and pregnancy), health service-related factors (e.g., geographical accessibility of the health facility), flexibility of PMCHT times with regard to the men’s occupational activities, waiting time at health facility, and participation in the ANC/PMCHT programs.

Male partner’s knowledge about PMCHT services was measured by using 10 items with the minimum and maximum scores of 0 and 10, respectively (12). The scoring of the answer choices (‘Yes’, ‘No’, and ‘Not known’) for item 1 (HIV-infected mother can transmit HIV to her child) was different from that of other items.

The reliability of the 10 items measuring knowledge on HIV/AIDS transmission was estimated rendering the Cronbach’s alpha coefficient of 0.72. Male partner participation in PMCHT programs was determined using male participation index obtained by using six items with equivalent weights (12). Some of the items were as follows:

• Did you know about your wife’s appointment for ANC the last time she was pregnant?
• Have you discussed with your female mate regarding the receipt of HIV counseling and testing in her previous pregnancy?

The response options were ‘yes’ and ‘no’ for each item. In this study, the Cronbach alpha coefficient for these items was obtained as 0.82.

Attitude was measured based on a 10-item 5-point Likert scale ranging from ‘strongly agree’ to ‘strongly disagree’. The participants’ total score in this scale was obtained by summing up the scores of 10 items, and the mean score was used to categorize the respondents (13). In this scale, individual response was rated in a range of 0-3 depending on the weight, and then added to yield the overall score ranging within 0-30.

In the current study, Cronbach alpha coefficient for the attitude scale was estimated as 0.8. The validity of the interview questionnaire was determined using the content validity. The questionnaires were considered as preliminary with regard to literature and the up-to-date similar findings. Data collection was accomplished using a modified, reviewed, pretested, and structured interview questionnaire adopted from the related literature (10-15). The data collection instrument primarily prepared in English was translated to native language (i.e., Afaan Oromo), and then retranslated to primary language via an independent translator to check for its consistency.

Before the actual interview, the ground was made smooth by the selection of a background with no or little disturbance, clarification of the objective of the interviewing, identification of the confidentiality terms, enlightenment of the interview contents, and display of the length of the interview. Actual interview was conducted by asking a single question at a moment, trying not to take sides, giving encouraging responses (e.g., occasional nods of the head and saying "uh huh"), and providing transition between major topics. 

The data were collected through structured interviews by nine trained nurses with a diploma degree who were fluent in Amharic and Afaan Oromoo languages. The interviews were conducted individually with the male spouses at their house. Data collection process was supervised by two nurses holding Bachelor's degree. To ensure the quality of the data, appropriate data collection tool was designed and pretested.

Pretesting was performed on 5% of the sample size in Dukem town as it is almost identical to the study setting. Data collectors and supervisors received a one-day training regarding the study objectives, data collection tools, and interview technique, as well as the ways of approaching the potential respondents and keeping confidentiality.

Data management, Analysis, and Interpretation

We manually checked for the completeness and consistency of the data. Data entry was accomplished using the Epi Data 3.1. Data analysis was performed in SPSS software (version 20.0). Descriptive statistics (e.g., percentages, mean, and standard deviation) were computed to summarize the characteristics of the study participants. In addition, bivariate analysis was run to select candidate variables for multivariable analysis. Factors with p-values less than 0.2 in bivariate model were selected for the final logistic regression model to determine the variables significantly associated with male partner participation in the PMCHT services. Furthermore, Pearson's Chi-squared test was employed to evaluate the variation across the categories of variables. Significance level for adjusted odds ratio was adjusted at 95% confidence interval with a p-value of less than 0.05.

Ethical Considerations

The institutional review board of Jimma University Institute of Health, Ethiopia, approved the study protocol. A permission letter was also obtained from the Health Bureaus of Oromia region and Health Administration Unit of Bishoftu town. Additionally, verbal consent was obtained from all participants.

Results

Sociodemographic profile of participants

A total of 405 male partners participated in the interview with a response rate of 96.2%. Nearly half of the male spouses (52.1%) were in the age group of 30-39 years with the mean age of 34.03±5.8 years. Regarding the monthly household income, 174 (43%) respondents were earning more than 3000.00 Ethiopian Birr (ETB) (Table 1).

                                         Table 1. Sociodemographic profiles of study population (n=405)

Information source, knowledge, and attitude-related characteristics of male partners

With regard to male partners’ information about the PMCHT services, 235 (58%) cases had heard about such programs. In terms of the information source, 89 (21.9%), 95 (23.4%), and 51 (12.6%) subjects had heard about this intervention from the media (i.e., radio and television), health center/hospital, and urban health extension workers, respectively.

With regard to male partners’ knowledge about PMCHT services, almost all 397 (98%) subjects knew that HIV can be transmitted from the infected mother to the child. However, only 316 (78%) participants knew that the utilization of PMCHT service prevents from PMCHT.  With respect to the attitude of the partners towards the PMCHT services, 237 (58.5%) respondents had negative attitudes toward these services. The minimum and maximum total scores of the respondents were 21 and 47, respectively, with the mean score of 36.1±4.7.

Level of Male Partner Participation in the prevention of mother-to-child HIV transmission programs

Out of the 405 interviewees, 214 (52.8%) cases did not attend to the centers delivering antenatal services with their partners. Furthermore, 207 (51.1%) male partners were tested and counseled for HIV, together with their wives, at clinics delivering ANC. Out of the total participants, 163 (40.2%) subjects were aware of the date of their spouses' antenatal appointment and 400 (98.8%) partners supported their spouses financially for ANC attendance (Table 2).

Determinants of partner involvement in the prevention of mother-to-child HIV transmission programs

During the bivariate analysis of partner involvement in the PMCHT services, eleven variables were selected for the final multivariate logistic regression (Table 3). Based on the multivariate logistic model, the significant predictors for partners’ involvement in the

Table 2. Male partners’ involvement status in antenatal care/prevention of mother-to-child HIV transmission programs (n=405)

ANC: antenatal care, PMCHT: prevention of mother-to-child HIV transmission, HCT: HIV counseling and testing

Table 3. Bivariate analysis of male partner involvement in the prevention of mother-to-child HIV transmission programs and the associated variables

PMCHT: prevention of mother-to-child HIV transmission, ANC: antenatal care, HCT: HIV counseling and testing

1-Reference category, * P < 0.05, ** P < 0.001

Table 4. Multivariable logistic regression model for the predictors of male partner involvement in the prevention of mother-to-child HIV transmission programs

PMCHT: prevention of mother-to-child HIV transmission, ANC: antenatal care

1-Reference category, * P < 0.05, ** P < 0.001